Relative Safety of Systemic Immunomodulatory Agents for Atopic Dermatitis

Atopic dermatitis on face
Atopic dermatitis on face
The risk of bacterial and opportunistic infection in patients with AD treated with systemic immunomodulatory agents is assessed.

The safety profiles of several systemic immunomodulatory agents in adults with atopic dermatitis (AD) were outlined in study data published in the Journal of the American Academy of Dermatology. In a population-based cohort of patients with AD, the incidence of serious infections was lower in those who received cyclosporine or methotrexate compared with those who received other immunomodulatory drugs.

Investigators conducted a population-based study using longitudinal claims data from the IBM MarketScan database for the years 2003 through 2017. The study enrolled adults with AD who were treated with systemic drugs. There were 5 systemic nonbiologic agents assessed in pairwise comparisons: methotrexate, cyclosporine, azathioprine, prednisone, and mycophenolate. The primary endpoint was development of a serious infection that led to hospital admission. Serious bacterial and opportunistic infections were identified using the associated International Classification of Disease diagnosis codes. In pairwise comparisons, propensity score matching was performed to minimize the effect of confounders. For each exposure pair, the relative risk (RR) for infection was computed.

The study cohort comprised as many as 232,611 eligible adults with AD. Among patients who used nonbiologic systemic immunomodulators (n=225,023), the incidence of serious infections was 7.53 (95% confidence interval [CI], 7.53-7.89) events per 1000 patients. Among patients who received phototherapy (n=7588) and dupilumab (n=391), the incidence rates were 7.38 (95% CI, 5.68-9.57) and 2.6 (95% CI, 0.45-14.3) events per 1000 patients, respectively. The 6-month incidence of serious infections after initiation of systemic nonbiologic agents ranged from 6.9 to 37.2 per 1000 patients. Compared with methotrexate (MTX), cyclosporine had a nominally reduced risk for serious infections (RR, 0.87; 95% CI, 0.59-1.29). The remaining nonbiologic agents had increased risk compared with MTX, with RRs of 1.78 (95% CI, 0.98-3.25), 1.89 (1.05-3.42), and 3.31 (1.94-5.64) for azathioprine, prednisone, and mycophenolate, respectively.

Based on this analysis of a commercially insured population, cyclosporine and MTX appeared to have similar associated infection risks. Prednisone, azathioprine, and mycophenolate, by comparison, were associated with elevated infection risk compared with MTX.

Study limitations include the small number of serious infections, which resulted in wide confidence intervals. In addition, as the period of interest was within 6 months of treatment initiation, long-term toxicity could not be assessed.

Despite these limitations, “[t]hese findings can help inform clinicians in their selection of medications for patients requiring systemic therapy for AD,” investigators wrote.

Disclosure: One study author declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of authors’ disclosures. 


Schneeweiss MC, Perez-Chada L, Merola JF. Comparative safety of systemic immunomodulatory medications in adults with atopic dermatitis. J Am Acad Dermatol. 2021;85(2):321-329. doi:10.1016/j.jaad.2019.05.073